Identi¢cation of novel Crenarchaeota and Euryarchaeota clusters
associated with di¡erent depth layers of a forest soil
Manuel Pesaro
a, Franco Widmer
b;a Swiss Federal Institute for Technology (ETH-Zu«rich), Institute of Terrestrial Ecology, Soil Biology, 8952 Schlieren, Switzerland
b Swiss Federal Research Station for Agroecology and Agriculture, Molecular Ecology (FAL-Reckenholz), Reckenholzstrasse 191, 8046 Zurich, Switzerland
Received 27 February 2002 ; received in revised form 30 May 2002 ; accepted 3 June 2002 First published online 20 July 2002
Abstract
Archaea have been shown to be ubiquitous among soil microbial communities. However, our knowledge on their diversity and spatial distribution in soil ecosystems is still limited. This study was conducted to investigate archaeal community changes along a forest soil depth profile in Unterehrendingen, Switzerland. From four consecutive soil depth layers, bulk soil DNA was extracted. Archaea-specific PCR amplification of small subunit ribosomal RNA genes (rDNA) was performed and combined with restriction fragment length polymorphism (RFLP) analysis with restriction endonuclease HaeIII [Bundt et al., Soil Biol. Biochem. 33 (2001) 729^738]. Significant changes of the RFLP fingerprints were reproducibly observed from the soil surface to 1 m depth. From the surface soil layer (0^9 cm) and the bottom soil layer (50^100 cm), libraries of PCR-amplified archaeal rDNA fragments were constructed. Screening of the libraries yielded various clones of different HaeIII RFLP types from the surface and the bottom soil layers, revealing shifts in major archaeal components along the soil depth profile. Clones of all RFLP types were sequenced and phylogenetically affiliated. These analyses revealed even more pronounced Archaea community shifts along the depth gradient. Several novel soil archaeal clusters were identified and some appeared predominantly associated to either the surface or the bottom soil layer. Euryarchaeal rDNA sequences, not yet reported from aerated soils, were found in the surface soil layer and were affiliated to the order Thermoplasmales and relatives. Novel crenarchaeal soil clusters were identified that included sequences only retrieved from the bottom soil layer. In this study, a this far unreported variety of archaeal groups was found in a forest soil ecosystem. The distinct depth-related community shift suggested the occurrence of different archaeal types that depend on environmental parameters that change along the soil depth profile. @ 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
Keywords : ARDRA ; Archaea community changes; Soil depth pro¢le ; Small subunit rDNA ; Phylogenetic analysis
1. Introduction
For many years it was believed that Archaea are asso-ciated with extreme environments only and that they rep-resent archaic life forms adapted to the harsh environmen-tal conditions that existed on earth billions of years ago
[1]. Initially, archaeal diversity was based on isolates and
has led to the di¡erentiation of two distinct lineages
termed Euryarchaeota and Crenarchaeota [2].
Euryar-chaeota appeared as a physiologically diverse group, which included extreme halophiles, thermophiles, and methanogens. Crenarchaeota exclusively included
sulfur-dependent hyperthermophiles [2]. Only in recent years
and by applying cultivation-independent molecular tools, it has become evident that Archaea are ubiquitous and abundant organisms, coexisting with other microorgan-isms in various environments. Assisted by the use of ge-netic markers, namely the small subunit (SSU) ribosomal RNA (rDNA) gene, the phylogeny of various isolates and
environmental SSU rDNA clones has been inferred [2];
see also the Ribosomal Database Project (RDP-II ; [3]).
This approach has led to the identi¢cation of novel un-cultivated archaeal clusters that clearly separate from their
cultivated relatives (for a review see DeLong [4]). Within
Euryarchaeota, novel groups have been described, which cluster with the diverse order Thermoplasmales and rela-tives. They appear to be important components of marine
picoplankton [5,6] and sediment communities [7,8], and
were also detected in the anoxic water columns of lakes
0168-6496 / 02 / $22.00 @ 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. * Corresponding author. Tel. : +41 (1) 377 73 76;
Fax : +41 (1) 377 7201.
E-mail address :[email protected](F. Widmer).
[9,10]. Non-aquatic representatives of Thermoplasmales have been retrieved from predominantly anoxic
environ-ments such as rice ¢eld soils[11,12], and the gut of a soil
feeding termite[13]. Many of the environmental sequences
cluster with the euryarchaeal orders Methanomicrobiales, Methanobacteriales, and Methanosarcinales, which have previously been isolated and cultured from the environ-ment. Crenarchaeal phylogeny has fundamentally changed by the isolation of novel SSU rDNA clones from meso-philic and psychromeso-philic habitats. Five new and distinct clusters of uncultivated Crenarchaeota have been derived from various non-thermophilic environments. Clones from
marine environments (water column [5,6,14,15]; sediment
[16]; the gut of a deep sea cucumber [17]) and lake
sedi-ments[18] form a prominent cluster referred to as Group
I.1a of uncultured Crenarchaeota [4]. A second cluster,
referred to as Group I.1b of uncultured Crenarchaeota
[4], is formed by clones that have been retrieved from
agricultural and forest soils [19^25] as well as from lake
sediments [26] and the gut of a soil feeding termite [13].
Sequences isolated from a boreal forest soil form a novel terrestrial soil cluster referred to as Group I.1c of
uncul-tured Crenarchaeota[4,22]. Group I.2of uncultured
Cren-archaeota has been inferred from few sequences isolated
from lake and marine sediments [4,16,26]. Group I.3 of
uncultured Crenarchaeota comprises sequences from lake
water [10], lake sediments [26,27], rice ¢eld soil[11], and
deep subsurface paleosol[28]. Despite all the e¡ort
under-taken for describing and characterizing the occurrence of Archaea in various environments, no isolates of non-ther-mophilic Crenarchaeota have been cultivated so far. This has hindered a biochemical and physiological character-ization of these still mysterious organisms.
A strategy to learn more about the ecological role of uncultivated Archaea is to study changes of their occur-rence in various habitats. This may help to identify Ar-chaea growth promoting environmental factors, and in turn may help to develop new isolation and cultivation strategies. For agricultural and forest soils, few studies have been performed to relate archaeal abundance or pop-ulation structures and spatial distributions. The interior and the surface of plant roots have been described as a niche with increased abundance and activity of certain
crenarchaeal groups [29,30].
In a recent study[31], we applied PCR ampli¢cation of
an Archaea-speci¢c SSU rDNA fragment followed by re-striction fragment length polymorphism (RFLP) analyses with restriction endonuclease HaeIII to resolve Archaea community structures along a pro¢le to 1 m depth of a forest soil in Switzerland. We reported gradual changes in Archaea community HaeIII RFLP ¢ngerprints along the depth pro¢le. In the present study the investigations were expanded by cloning these archaeal SSU rDNA fragments and by sequence analyses. The objective of this study was to identify the Archaea community components contribut-ing to the observed archaeal HaeIII RFLP ¢ngerprint
changes between the surface and the bottom soil layers
in the depth pro¢le[31]. Our hypothesis was that changes
of soil characteristics along the depth pro¢le, as revealed by chemical parameters, promoted growth of phylogeneti-cally di¡erent archaeal groups.
2. Materials and methods
2.1. Field site and experimental design
The ¢eld site for this study has been described in detail
by Bundt et al. [31]. Brie£y, the experimental forest plot
was located near Unterehrendingen, Switzerland. The stand was planted in 1930 dominantly with Norway spruce (Picea abies (L.) Karst.) mixed with beech (Fagus sylvatica (L.)) and few other species. The soil type in the area was a
typical Haplumbrept[32].
2.2. Soil sampling and analyses
Soil samples for DNA analyses were taken in October 1998 in the context of a study to investigate the e¡ects of preferential water £ow on biological characteristics in soil
[31]. For this purpose 45 mm water stained with the food
dye Brilliant blue (C.I. 42090) (3 g l31) was sprinkled
during 6 h onto a 1U1.5 m wide area. This simulated a rainfall event and stained preferential water £ow paths in
the soil[33]. One day after application of the dye solution,
a trench was opened to 1.2m depth. A vertical soil pro¢le of 1U1 m was prepared and divided into depth layers of 0-9 cm, 9-20 cm, 20-50 cm, and 50-100 cm. The blue-stained areas were de¢ned as preferential water £ow paths, the non-stained areas as soil matrix. Three types of soil samples were collected, i.e. (1) preferential £ow path soil from the blue-stained areas ; (2) matrix soil from the un-stained areas ; and (3) bulk soil samples collected from the whole area of each depth layer. The samples used for de-tailed molecular analyzes in the present study were the bulk soil samples, which represented the regular soil type and included preferential £ow path areas and matrix soil. In the trench, a series of ¢ve consecutive soil pro¢les was prepared at 10-cm intervals. Soil samples from corre-sponding depth layers of all ¢ve consecutive soil pro¢les were pooled and mixed. Chemical and physical soil
anal-yses were described by Bundt et al. [31](Table 1).
2.3. DNA extraction
DNA was extracted from fresh soil samples using a bead beating procedure. Brie£y, fresh soil (approximately 0.5 g dry wt. equivalent) was added to 1.5-ml reaction tubes containing 0.5 g glass beads (0.1 mm diameter). 1 ml extraction bu¡er (100 mM Tris/HCl, pH 7.4, 10 mM EDTA, 1.5% SDS, 1% deoxycholate, 1% Nonidet
added to each sample. Bead beating was performed for 1 min with a cell homogenizer (MSK Zellhomogenisator, B. Braun Biotech International, Melsungen, Germany) at
4000 oscillations min31. Slurries were centrifuged (1 min,
16 000Ug) and supernatants were saved and extracted once with 1 volume phenol/chloroform/isoamylalcohol (25/24/1) and twice with 1 volume chloroform/isoamylal-cohol (24/1). Extracts were mixed with 1 volume precip-itation solution (20% polyethylene glycol 6000, 2.5 M NaCl), incubated at 37‡C for 1 h and centrifuged at room temperature (15 min, 16 000Ug). Pellets were washed once with 0.5 ml 70% ethanol, air dried, and re-suspended in TE-bu¡er (10 mM Tris/HCl, 1 mM EDTA,
pH 8) at 1 ml (g dry wt. soil extracted)31 [35]. Extracted
DNA was quanti¢ed using PicoGreen0 (Molecular
Probes, Eugene, OR, USA) according to the procedure of Sandaa et al.[36].
2.4. PCR ampli¢cation and RFLP analyses
2ng soil DNA was used for each PCR ampli¢cation
performed according to Widmer et al.[37]. For PCR, an
Archaea-speci¢c forward-primer (ARCH915-for : 5P-AG-GAATTGGCGGGGGAGCAC-3P; Escherichia coli
num-bering 915^934 bp[38]) targeting the SSU rRNA genes in
conjunction with a universal SSU rRNA gene reverse primer (UNI-b-rev : 5P-GACGGGCGGTGTGT(A/G)C-AA-3P; E. coli numbering 1390^1407 bp) modi¢ed [31]
from Amann et al. [38] were used. PCR ampli¢cation
was performed with 1 U Taq DNA polymerase (Amer-sham, Zu«rich, Switzerland) and the supplied bu¡er with
2mM MgCl2. The annealing temperature was 65‡C and
PCR was performed with 40 ampli¢cation cycles [31].
PCR product quality was analyzed on 1.2% UltraPure agarose gels (Gibco/BRL, Life Technologies AG, Basel, Switzerland). RFLP analysis was performed using restric-tion endonuclease HaeIII (Boehringer Mannheim,
Rot-kreuz, Switzerland) according to Widmer et al. [39].
RFLP patterns were analyzed by electrophoresis in 4% MetaPhor gels (FMC BioProducts, Rockland, USA) and ethidium bromide staining. Gels were photographed using Polaroid 677 ¢lms (Polaroid, Uxbridge, UK). Images were scanned (600 dpi Apple ColourOne scanner), and inten-sities of distinct RFLP bands from each pattern were quanti¢ed using NIH Image v.1.61 according to Widmer
et al. [40]. For statistical analyses of single bands, total
lane intensities (de¢ned as the sum of the intensities of
all bands detected in one lane) was set to 100% and each band was expressed as percentage of this standar-dized sum. Two-sided t-tests for pairwise comparisons of standardized band intensities were performed using Excel 98 (Microsoft Corporation, Redmond, WA, USA). Linear correlations between band intensities in community HaeIII RFLP ¢ngerprints and band occurrences in the gene libra-ries were calculated with Excel 98 (Microsoft Corpora-tion).
2.5. DNA cloning and sequencing
PCR products were cloned without prior puri¢cation
using the pGEM0-T Easy cloning kit (Promega, Madison,
WI, USA) and E. coli JM109 (Promega). Gene libraries were screened by touching white bacterial colonies with a pipette tip and adding cells to 50 Wl PCR ampli¢cation mixes, prepared as described above for the Archaea-spe-ci¢c PCR. Ampli¢cation was performed with 30 ampli¢-cation cycles using the conditions described above. PCR products were analyzed on 1.2% UltraPure agarose gels and positive products were subjected to RFLP analyses as described above. Clones were classi¢ed based on their HaeIII restriction patterns and labeled ‘a’, ‘b’, ‘c’, etc. Rarefaction analysis of HaeIII restriction patterns was performed for each library with the program Resampling
Rarefaction 1.0 (http ://www.uga.edu/Vstrata/software/)
using 104 resamplings. Maximal clone richness in each
library was estimated by parametric calculation of the asymptote from rarefaction curves using the Michaelis^
Menten equation [41]. Plasmid DNA of representative
clones was puri¢ed using Wizard plasmid miniprep col-umns (Promega). The sequences of the approximately 495 bp long cloned Archaea SSU rDNA PCR products (location according E. coli numbering : 915^1407 bp) were determined from both strands (T7 and SP6 sequencing primer sites) using an ABI310 Genetic Analyzer (Applied Biosystems, Rotkreuz, Switzerland).
2.6. DNA sequence analyses
The RFLP patterns of cloned archaeal SSU rDNA frag-ments were con¢rmed using sequence-based calculated re-striction fragmentation with MacDNASIS v. 3.6 (Hitachi Software Engineering America, Ltd., San Bruno, CA, USA). Calculated fragmentation patterns were presented as fragment sizes on a logarithmic scale. Twenty new clone
Table 1
Physical and chemical soil characteristics of the four forest soil depth layers
Depth [cm] pHa Sand [%] Silt [%] Clay [%] N
tot[%] Corg[%] 0^9 3.4 35.8 45.4 18.8 0.137 2.565 9^20 3.7 35.9 45.6 18.5 0.050 0.855 20^50 3.8 34.5 45.6 19.9 6 0.03 0.471 50^100 3.9 29.4 48 22.6 6 0.03 0.288 apH in 10 mM CaCl 2.
sequences and 118 control sequences retrieved from the
public domain databases RDP-II (http ://rdp.cme.msu.edu)
and GenBank (www.ncbi.nlm.nih.gov/GenBank/) were
aligned using the BioEdit version 5.0.9 sequence analysis
software [42]. Phylogenetic analyses were performed with
BioEdit version 5.0.9 expanded with selected Phylip
soft-ware package modules[43]. Sequence distance-based
phy-logenetic tree inference was performed on the whole DNA sequence of the PCR products with JukespCantor
dis-tance value calculation [44] and neighbor joining cluster
analysis[45]. For maximum likelihood inference of
phylo-genetic relationships the fastDNAml program [46] was
used. Inferred trees were viewed and edited using Tree Explorer version 2.12 from the MEGA2 software package
[47].
2.7. Nucleotide sequence accession numbers
Nucleotide sequences of the 20 novel Archaea SSU rDNA clones determined in this study have been deposited in the GenBank database with the accession numbers AF458627^AF458646.
3. Results
3.1. DNA extraction and detection of Archaea populations Image analysis revealed that the three di¡erent soil sam-ple types, i.e. preferential water £ow path soil, matrix soil, and bulk soil, all displayed the same depth-dependent changes in the Archaea HaeIII RFLP ¢ngerprints [31] (Table 2). Therefore the bulk soil samples from each depth layer, representing the actual soil type including preferen-tial £ow paths and matrix soil, were chosen for detailed cloning and sequencing analyses of archaeal SSU rDNA PCR products.
The quantities of DNA extracted from fresh bulk soil markedly decreased with increasing depth in the pro¢le. In the surface soil layer (0^9 cm), DNA quantity was highest
with 30 Wg g31. Between 9 and 20 cm, 24 Wg g31 DNA was
extracted. DNA quantity dropped to 13 Wg g31 in the soil
layer between 20 and 50 cm and was lowest in the bottom
soil layer (50^100 cm) with 3 Wg g31. PCR ampli¢cation of
target rRNA gene fragments was performed on the same quantity (2ng) of bulk soil DNA to allow for a direct comparison of relative band intensities in the HaeIII
RFLP ¢ngerprints (Fig. 1). High-resolution agarose gel
analysis resolved ¢ve prominent bands within the Archaea
community HaeIII RFLP ¢ngerprints (Fig. 1, bands I^V).
Densitometric quanti¢cation and statistical analysis of band intensities from bulk soil, preferential water £ow path soil, and matrix soil revealed signi¢cant changes in the HaeIII RFLP ¢ngerprints between the four depth
layers ([31]; Table 2). Intensity of Band I decreased
from the surface soil layer (0^9 cm) to the bottom soil layer (50^100 cm) to 47% (**P 6 0.01). Band III displayed the opposite trend as it increased in intensity from the surface to deeper soil layers with a maximum of 170% in the bottom relative to the surface soil layer (*P 6 0.05). Band II was only detected in the surface, but not in deeper soil layers. Band IV revealed no signi¢cant changes along the entire depth pro¢le, whereas band V showed a slight increase to 124% intensity relative to the surface soil layer (*P 6 0.05). No signi¢cant di¡erences were observed when comparing total lane intensities between the four depth layers.
Fig. 1. HaeIII RFLP patterns of archaeal 16S rDNA fragments ampli-¢ed from bulk soil DNA extracts from depth layers 0^9, 9^20, 20^50 and 50^100 cm of a Swiss forest soil. MW : 1 kb molecular mass marker (Promega) ; (3): negative control. The arrowhead indicates the migra-tion posimigra-tion of the undigested Archaea PCR-product at approximately 500 bp.
Table 2
Relative band intensitiesaof Archaea community RFLP ¢ngerprints along the depth gradient
Band labelb Depth layer [cm]
0^9 9^20 20^50 50^100 I 1.00 T 0.05 1.17 T 0.020.82T 0.01 0.47 T 0.05 II 1.00 T 0.25 NDc NDc NDc III 1.00 T 0.14 1.41 T 0.07 1.65 T 0.05 1.70 T 0.01 IV 1.00 T 0.07 0.94 T 0.01 1.05 T 0.01 1.17 T 0.03 V 1.00 T 0.04 1.04 T 0.01 1.10 T 0.01 1.24 T 0.02
aIntensities of bands were determined densitometrically and were expressed relative to the value determined for the respective bands detected in the
sur-face soil layer. Standard deviations were calculated using the three soil sample types from each depth layer as replicates.
bRFLP bands labeled ‘I’^‘V’ according to labeling indicated inFig. 1. cND : not detectable.
3.2. Characterization of changing Archaea populations In order to gain detailed information on the Archaea populations represented by the changing HaeIII RFLP ¢ngerprints in the di¡erent soil layers, gene libraries of Archaea SSU rDNA ampli¢ed from the surface soil layer (0^9 cm) and the bottom soil layer (50^100 cm) were con-structed. The two libraries were screened by use of Ar-chaea-speci¢c PCR and HaeIII RFLP analysis performed on single clones. Among the 104 clones screened (39 from the surface soil library and 65 from the bottom soil li-brary), eight di¡erent HaeIII RFLP patterns were
ob-served (Fig. 2). The abundance of each pattern was
quan-ti¢ed in both gene libraries, which allowed for a numeric comparison of the relative clone compositions of the top
and the bottom soil libraries (Table 3). Richness
estima-tions indicated that from the surface soil library 91% (5 of 5.5) and from the bottom soil library 89% (5 of 5.6) of the patterns were recovered (data not shown). HaeIII RFLP pattern ‘a’ was 1.4-fold more abundant in the bottom soil library while patterns ‘b’, ‘c’ and ‘e’ were exclusively found in the surface soil library. Pattern ‘d’ was 7.7-fold more abundant in the bottom soil library. Patterns ‘f’^‘h’ were infrequent and restricted to the bottom soil library. Fur-ther analysis revealed that each of the ¢ve prominent bands observed in the complex Archaea community
RFLP ¢ngerprints (Fig. 1) was also identi¢ed in the
RFLP ¢ngerprint of at least one clone (Fig. 2 andTable
4). Band I was exclusively found in pattern ‘b’ while band
II was detected in pattern ‘e’ only. Band III was mainly attributed to pattern ‘d’ with a minor contribution of pat-tern ‘f’, while band IV mainly originated from patpat-terns ‘a’ and ‘c’ with minor contributions of patterns ‘g’ and ‘h’. Band V was composed by patterns ‘a’^‘e’ with minor con-tributions of patterns ‘f’ and ‘g’. The R2-value of the linear
correlation between band intensities in the community
HaeIII RFLP ¢ngerprints and the band occurrences in the gene libraries was 0.54. These analyses revealed that each of the ¢ve HaeIII RFLP bands displayed the same trend of abundance in the two clone libraries as indicated by the band intensities of the Archaea community HaeIII RFLP ¢ngerprints.
3.3. Phylogenetic analyses of Archaea clones
The question of which phylotypes were represented in the Archaea community HaeIII RFLP ¢ngerprints was addressed by sequencing Archaea SSU rDNA clones, which represented the di¡erent HaeIII RFLP types. The 20 new sequences isolated from the surface and the bot-tom layers of the forest soil depth pro¢le were aligned to 118 de¢ned control sequences derived from public data-bases. Only control sequences that covered the entire SSU rDNA region de¢ned by the PCR primers used in this study were included in the analysis. The phylogeny of the aligned clone sequences was inferred by using standard distance estimation and cluster analysis routines (data not
shown) as well as a maximum likelihood routine (Fig. 3).
The resulting phylogenetic trees revealed similar tree to-pologies and in particular identical clustering of the
dis-tinct clusters ‘A’^‘E’ indicated in Fig. 3.
Table 3
RFLP pattern frequencies in archaeal SSU rDNA fragment libraries RFLP pattern labelsa Archaea SSU rDNA libraries
surface soilb bottom soilb
ac 48.7 (19) 67.8 (44) b 17.9 (7) ^ c 23.1 (9) ^ dc 2.6 (1) 20.0 (13) e 7.6 (3) ^ f ^ 3.1 (2) g ^ 1.5 (1) h ^ 1.5 (1) x ^ 6.2(4)
Total 100.0 (39 clones) 100.0 (65 clones)
aRFLP pattern labels as de¢ned in Fig. 2; ‘x’ represents cloning
arti-facts.
bRFLP pattern frequencies are presented in percent for each library.
The number of clones identi¢ed is given in parentheses.
cSequence and phylogenetic analyses (Fig. 3) revealed that clones from
the surface and bottom soil layers, which shared the same HaeIII RFLP pattern, belonged to di¡erent phylotypes.
Fig. 2. Calculated HaeIII RFLP patterns that occurred among 104 cloned archaeal 16S rDNA fragments from surface (0^9 cm) and bot-tom (50^100 cm) soil layers from a Swiss forest soil. Fragment sizes were inferred from DNA sequences of representative clones. MW : 1 kb molecular mass marker (Promega) ; RFLP patterns were labeled ‘a’^‘h’. Bands I^V were detected in the complex RFLP ¢ngerprints (see also
Fig. 1aandTable 3).
Table 4
Assignment of individual RFLP bands in Archaea community ¢nger-prints to speci¢c RFLP patterns
Banda Sizeb RFLP patternc
I 355 b
II 269 e
III 218 d (f)d
IV 184 a, c, (g, h)d
V 138 a, b, c, d, e, (f, g)d aBand labels as de¢ned inFig. 1.
bFragment sizes (in bp) of bands I^V. Sizes were determined from
se-quence data as shown inFig. 2.
cRFLP pattern labels as de¢ned inFig. 2.
Fig. 3. Phylogenetic tree based on maximum likelihood calculation. Each of the archaeal SSU rDNA sequences is identi¢ed by its clone name, by its source, and by its sequence accession number. Sequences isolated in the present study are printed in bold. Clone names indicate the origin of the di¡er-ent clones, i.e. layer (surface soil : 03 ; bottom soil : 12) followed by the clone number (01^58) and the HaeIII RFLP type (‘a’ to ‘h’). Cluster (I) de¢nes the archaeal kingdom of Korarchaeota, cluster (II) the kingdom of Euryarchaeota, and cluster (III) the kingdom of Crenarchaeota. Shaded clusters ‘A’^‘E’ mark novel archaeal forest soil clusters. Numbers at condensed clusters indicate the number of sequences included. Brackets at the right margin mark phylogenetic clusters de¢ned in the literature[4]. The scale bar indicates the average substitution rate per nucleotide position.
Of the 20 sequences that were derived from the soil depth pro¢le described in this study, 16 grouped with Crenarchaeota and four with Euryarchaeota. The cren-archaeal sequences were associated with four distinct clus-ters (‘A’^‘D’). Cluster ‘A’ included clones 1211d and 12 19d from the bottom soil layer and displayed highest sim-ilarity to two sequences from deep subsurface acidic mine
water (clones SAGMA-D and SAGMA-X ; [48]) and to
one sequence from groundwater (clone SRS62DAR03). Cluster ‘B’ included four sequences from the surface soil (clones 03 01a, 03 02a, 03 03a, and 03 12a) and two se-quences from the bottom soil layer (clones 1228g and 12 58h). They grouped with four clones isolated from a wet-land soil in Japan (clones OS-19, OS-25, OS-31, and WSB-6). Two sequences from the bottom soil layer (clones 12 45f and 1247f) formed cluster ‘C’, which was only weakly associated with clone pJP96 from a Yellowstone hot spring. Three sequences from the bottom soil layer (clones
1201a, 1202a, and 1230a) formed cluster ‘D’, which
branched closely from a group containing eight clones from the Japanese wetland soil mentioned above (clones OS-6, OS-14, OS-21, OS-22, OS-26, OS-27, AM-11, and WSB-20). Two sequences from surface soil (clones 03 06c and 03 17c) were allocated to the terrestrial cluster of uncultured Crenarchaeota (Group I.1b) in close vicinity to sequences cloned from plant roots and agricul-tural bulk soil (several of the TRC-clones and clone SCA1145). One sequence retrieved from surface soil (clone 03 21d) formed a separate branch without close associa-tion to any sequences reported to date. Euryarchaeal se-quences derived from this study clustered exclusively with Thermoplasmales and relatives (clones 03 11b, 03 15b, 03 14e, and 03 27e) and formed a cluster termed ‘E’ within this diverse order. One sequence from Japanese wetland soil (clone OS-10) and one sequence from a
hydrocarbon-contaminated aquifer (clone WCHD3-16 ; [49]) supported
this cluster.
4. Discussion
Archaea community shifts along the depth pro¢le of a Swiss forest soil were recently reported by Bundt et al. [31] by use of HaeIII RFLP analysis of SSU rDNA PCR products. Little information is available on associations of speci¢c Archaea populations with changing chemical, physical, or biological conditions in soils. Such informa-tion, however, may allow to learn more about physiology
and potential functions of uncultivated Archaea [24,25].
The observed archaeal community shifts along a soil depth pro¢le, therefore, represented an interesting model system for assessing di¡erences in Archaea populations in closely connected natural soil habitats. In the present study, tailed molecular analyses were applied for a profound de-scription of the observed changes in archaeal community HaeIII RFLP ¢ngerprints between the surface soil layer
(0^9 cm depth) and the bottom soil layer (50^100 cm depth).
Archaea have repeatedly been described in various
agri-cultural and forest soils [19^25], however, until now no
isolates have been cultivated from soil systems. Therefore molecular analyses based on SSU rRNA and its genes can provide important contributions for a detailed assessment of archaeal communities in soil systems. The restrictions to this approach, however, are that only limited informa-tion on physiology and potential funcinforma-tions of the organ-isms can be derived from rRNA sequences. For Archaea this has been particularly limiting, since described archaeal physiologies have tight associations to extreme
environ-ments[2]. Therefore, it appears of great interest to assign
the occurrence of novel, uncultured Archaea populations to speci¢c environments and to gain knowledge about their preferred habitats and growth conditions.
The occurrence of sequences associated with Thermo-plasmales and relatives, an euryarchaeal order, was partic-ularly interesting, since no euryarchaeal sequences have yet been described in forest soils. In terrestrial habitats, Euryarchaeota have been found mostly in anaerobic
envi-ronments such as rice ¢eld soils [11,12,50,51] or the
me-thanogenic zone of lake sediments [52]. Therefore, it may
be surprising that in the forest soil depth pro¢le, Euryar-chaeota were only found in the surface layer, which was certainly better supplied with oxygen than the layer at 50^ 100 cm depth. RFLP analyses of single clones con¢rmed this depth-related occurrence. Sequences with HaeIII
re-striction patterns ‘b’ and ‘e’ (Fig. 2) revealed euryarchaeal
origin and displayed a high relative abundance in the
sur-face soil (18% and 7.6% ; Table 3). They were not found
among the 65 clones isolated from the bottom soil layer. Even though no detailed information on redox conditions were recorded from this experiment, the presence of se-quences from mesophilic Euryarchaeota in the surface soil layer of this forest soil suggested that certain groups of Thermoplasmales and relatives are not strictly depen-dent on anoxic environments. Buckley et al., 1998 [23] proposed that some terrestrial Crenarchaeota may be tol-erant to oxic conditions. Based on our observations we speculate that also some mesophilic Euryarchaeota may be tolerant to oxic conditions in soil. The recent identi¢-cation of a euryarchaeal sequence retrieved from maize
root [30], presumably a typical oxic habitat, supports
this hypothesis.
The novel crenarchaeal clusters ‘B’ and ‘D’, both con-tained clones with HaeIII RFLP pattern ‘a’. Interestingly, cluster ‘B’ contained only HaeIII RFLP pattern ‘a’-clones from the surface soil, while cluster ‘D’ contained HaeIII RFLP pattern ‘a’-clones from the bottom soil layer (Fig. 3 and Table 3). The two sequences from the bottom soil layer that displayed HaeIII RFLP pattern ‘d’ formed a novel soil cluster between Group I.1a and Group I.1b of uncultivated Crenarchaeota. The ¢rst group consists of sequences from marine environments and freshwater lake
sediments [4], whereas the second group is exclusively
composed of sequences from terrestrial habitats [4]. Only
one clone, which displayed HaeIII RFLP pattern ‘d’, was
isolated from the surface soil layer (Table 3) and was
phylogenetically separated from the same HaeIII RFLP types retrieved from the bottom soil layer. Thus, the marked increase in relative abundance of pattern ‘d’ from the surface to the bottom soil layer rather re£ected a change of di¡erent phylotypes with the same HaeIII RFLP pattern than an increase of a speci¢c phylotype
from the surface to the bottom layer (Table 3). These
di¡erent phylogenetic associations of clones sharing the same HaeIII RFLP patterns, i.e. patterns ‘a’ and ‘d’,
de-rived from the surface and bottom soil layers (Fig. 3)
indicated that the two soil layers harbored highly distinct archaeal populations. Although very powerful for answer-ing our research objectives, these ¢ndanswer-ings clearly indicated limitations of RFLP analyses with one or few restriction enzymes. We demonstrated that organisms of di¡erent phylogenetic a⁄liations shared the same HaeIII RFLP pattern and only could be distinguished by use of more
detailed phylogenetic analyses (Fig. 3).
Crenarchaeal organisms represented by RFLP pattern ‘c’ were exclusively identi¢ed in the surface soil layer (23% abundance) and clustered with Group I.1b of uncul-tivated terrestrial Crenarchaeota. The relatively minor as-sociation of forest soil derived archaeal sequences with this cluster was surprising, since the majority of clones
re-trieved from aerated soils fall into this group [19^
21,23,25,29]. The only exceptions are sequences retrieved from a Finnish forest soil, which form a unique cluster
termed FFSB [22]. Unfortunately, it was not possible to
relate our sequences to the FFSB cluster since no over-lapping sequence portions of the SSU rRNA genes were ampli¢ed by the two di¡erent PCR protocols. The phylo-genetic association of three sequences (clones 03 21d, 12 45f, and 1247f) within the kingdom of Crenarchaeota remained unclear due to the lack of related sequences in public databases. The closest relatives found in databases were clone SCA1158 with 86% sequence identity to clone 03 21d and clone pJD96 with 91% sequence identity to clones 1245f and 1247f. Checking the sequences for chi-meric characteristics indicated that they did not represent PCR artifacts of known SSU rRNA genes (data not shown). Therefore, it was not possible in this study to propose a con¢rmed phylogenetic a⁄liation for these three clones.
Based on chemical and physical analyses of the four soil
depth layers (Table 1), factors that correlated with
archae-al community changes could be identi¢ed. Organic carbon and total nitrogen showed particularly strong decreases and therefore paralleled the disappearance of Euryar-chaeota and Group I.1b CrenarEuryar-chaeota along the soil depth pro¢le. Fresh organic material has been related to increased archaeal abundance and methane production in
anaerobically incubated soil cores[53]. Conversely, texture
and pH exhibited less pronounced changes and may be of minor importance for the observed Archaea community shifts.
The approach of isolating and characterizing SSU rRNA gene sequences has been powerful for describing the diversity of the soil micro£ora. Since no archaeal iso-lates have been cultivated from soil, we still lack informa-tion on their physiology and funcinforma-tions. New approaches are being developed and used for gaining a better under-standing of these soil organisms. Information on their functional properties can be gained from metagenome
analyses [54,55]. With these approaches, high molecular
mass genomic DNA has been cloned directly from envi-ronmental habitats. Large inserts containing the rRNA operon allowed ¢rst to relate the cloned DNA to its archaeal source and second to explore the genome of the organism. Detailed studies on genome architecture of two marine Crenarchaeota have already been performed and
several protein-coding genes have been described [54,55].
It has been possible, for example, to express a DNA poly-merase from the native genes of the uncultured psychro-philic Crenarchaeum C. symbiosum. Subsequent biochem-ical characterization demonstrated its heat labile character
[56]. This ¢nding was therefore in agreement with the
psy-chrophilic phenotype inferred from the distribution of rRNA genes in the environment. The archaeal SSU rRNA gene sequences described in this study and else-where therefore may represent the keys to genomic analy-ses of uncultivated Archaea in soil.
Data obtained with the PCR/HaeIII RFLP approach in conjunction with phylogenetic analyses supported the hy-pothesis that distinct archaeal populations were present in di¡erent depth layers of a forest soil. Representatives of Euryarchaeota, associated with the order Thermoplas-males, were detected in the surface soil layer, a habitat were they have not been described before. Di¡erent archaeal populations were detected in either the surface or the bottom soil layer, indicating dependency on speci¢c environmental factors. Their roles in non-extreme environ-ments still remain unclear, however, growing knowledge on the association of speci¢c Archaea populations with well-characterized eco-niches may be useful to design cul-tivation-based approaches to retrieve them. In addition to serve as phylogenetic markers, speci¢c SSU rDNA sequen-ces, as described in this study, may be helpful for recover-ing uncultivated organisms or serve as anchors in metage-nome analyses.
Acknowledgements
The authors are grateful to Helmut Bu«rgmann (ETH Zurich) for help with phylogenetic sequence analyses. Prof. Josef Zeyer and Dr. Von Sigler (ETH Zurich) are acknowledged for continuous support and valuable com-ments on the manuscript. We thank Dr. Maja Bundt and
Dr. Peter Blaser (WSL Birmensdorf) for providing soil samples. M.P. was supported by funding from the Swiss Federal O⁄ce for Education and Science (COST Action 831).
References
[1] Woese, C.R., Magrum, L.J. and Fox, G.E. (1978) Archaebacteria. J. Mol. Evol. 11, 245^252.
[2] Woese, C.R., Kandler, O. and Wheelis, M. (1990) Towards a natural system of organisms : proposal for the domains Archaea, Bacteria and Eucarya. Proc. Natl. Acad. Sci. USA 87, 4576^4579.
[3] Maidak, B.L., Cole, J.R., Lilburn, T.G., Parker Jr., C.D., Saxman, P.R., Stredwick, J.M., Garrity, G.M., Li, B., Olsen, G.J., Pramanik, S., Schmidt, T.M. and Tiedje, J.M. (2000) The RDP (Ribosomal Database Project) continues. Nucleic Acids Res. 28, 173^174. [4] DeLong, E.F. (1998) Everything in moderation: Archaea as
‘non-extremophiles’. Curr. Opin. Genet. Dev. 8, 649^654.
[5] DeLong, E.F., Wu, K.Y., Prezelin, B.B. and Jovine, R.V.M. (1994) High abundance of Archaea in antarctic marine picoplankton. Na-ture 371, 695^697.
[6] Fuhrman, J.A. and Davis, A.A. (1997) Widespread Archaea and novel Bacteria from the deep sea as shown by 16S rRNA gene se-quences. Mar. Ecol.-Prog. Ser. 150, 275^285.
[7] Munson, M.A., Nedwell, D.B. and Embley, T.M. (1997) Phylogenetic diversity of Archaea in sediment samples from a coastal salt marsh. Appl. Environ. Microbiol. 63, 4729^4733.
[8] Hinrichs, K.-U., Hayes, J.M., Sylva, S.P., Brewer, P.G. and DeLong, E.F. (1999) Methane-consuming Archaebacteria in marine sediments. Nature 398, 802^805.
[9] Cytryn, E., Minz, D., Oremland, R.S. and Cohen, Y. (2000) Distri-bution and diversity of Archaea corresponding to the limnological cycle of a hypersaline strati¢ed lake (Solar Lake, Sinai, Egypt). Appl. Environ. Microbiol. 66, 3269^3276.
[10] Jurgens, G., Glo«ckner, F.O., Amann, R., Saano, A., Montonen, L., Likolammi, M. and Mu«nster, U. (2000) Identi¢cation of novel Ar-chaea in bacterioplancton of a boreal forest lake by phylogenetic analysis and £uorescent in situ hybridization. FEMS Microbiol. Ecol. 34, 45^56.
[11] Grosskopf, R., Stubner, S. and Liesack, W. (1998) Novel euryar-chaeotal lineages detected on rice roots and in the anoxic bulk soil of £ooded rice microcosms. Appl. Environ. Microbiol. 64, 4983^ 4989.
[12] Lueders, T. and Friedrich, M. (2000) Archaeal population dynamics during sequential reduction processes in rice ¢eld soil. Appl. Environ. Microbiol. 66, 2732^2742.
[13] Friedrich, M.W., Schmitt-Wagner, D., Lueders, T. and Brune, A. (2001) Axial di¡erences in community structure of Crenarchaeota and Euryarchaeota in the highly compartimentalized gut of the soil feeding termite Cubitermes orthognathus. Appl. Environ. Microbiol. 67, 4880^4890.
[14] Fuhrman, J.A., McCallum, K. and Davis, A.A. (1992) Novel major archaebacterial group from marine plankton. Nature 356, 148^149. [15] DeLong, E.F. (1992) Archaea in coastal marine environments. Proc.
Natl. Acad. Sci. USA 89, 5685^5689.
[16] Vetriani, C., Jannasch, H.W., MacGregor, B.J., Stahl, D.A. and Rey-senbach, A.-L. (1999) Population structure and phylogenetic charac-terization of marine benthic Archaea in deep-sea sediments. Appl. Environ. Microbiol. 65, 4375^4384.
[17] McInerney, J.O., Wilkinson, M., Patching, J.W., Embley, T.M. and Powell, R. (1995) Recovery and phylogenetic analysis of novel archaeal rRNA sequences from a deep-sea deposit feeder. Appl. En-viron. Microbiol. 61, 1646^1648.
[18] MacGregor, B.J., Moser, D.P., Alm, E.W., Nealson, K.H. and Stahl,
D.A. (1997) Crenarchaeota in Lake Michigan sediment. Appl. Envi-ron. Microbiol. 63, 1178^1181.
[19] Ueda, T., Suga, Y. and Matsuguchi, T. (1995) Molecular phyloge-netic analysis of a soil microbial community in a soybean ¢eld. Eur. J. Soil Sci. 46, 415^421.
[20] Bintrim, S.B., Donohue, T.J., Handelsman, J., Roberts, G.P. and Goodman, R.M. (1997) Molecular phylogeny of Archaea from soil. Proc. Natl. Acad. Sci. USA 94, 277^282.
[21] Bornemann, J. and Triplett, E.W. (1997) Molecular microbial diver-sity in soils from eastern Amazonia : evidence for unusual microor-ganisms and microbial population shifts associated with deforesta-tion. Appl. Environ. Microbiol. 63, 2647^2653.
[22] Jurgens, G., Lindstrom, K. and Saano, A. (1997) Novel group within the kingdom Crenarchaeota from boreal forest soil. Appl. Environ. Microbiol. 63, 803^805.
[23] Buckley, D.H., Graber, J.R. and Schmidt, T.M. (1998) Phylogenetic analysis of nonthermophilic members of the kingdom Crenarchaeota and their diversity and abundance in soils. Appl. Environ. Microbiol. 64, 4333^4339.
[24] Jurgens, G. and Saano, A. (1999) Diversity of soil Archaea in boreal forest before, and after clear-cutting and prescribed burning. FEMS Microbiol. Ecol. 29, 205^213.
[25] Sandaa, R.A., Enger, O. and Torsvik, V. (1999) Abundance and diversity of Archaea in heavy-metal-contaminated soils. Appl. Envi-ron. Microbiol. 65, 3293^3297.
[26] Hershberger, K.L., Barns, S.M., Reysenbach, A.L., Dawson, S.C. and Pace, N.R. (1996) Wide diversity of Crenarchaeota. Nature 384, 420.
[27] Schleper, C., Holben, W. and Klenk, H.P. (1997) Recovery of crenar-chaeotal rDNA sequences from freshwater lake sediments. Appl. En-viron. Microbiol. 63, 321^323.
[28] Chandler, D.P., Brockmann, F.J., Bailey, T.J. and Fredrickson, J.K. (1998) Phylogenetic diversity of Archaea and Bacteria in deep sea subsurface paleosol. Microb. Ecol. 36, 37^50.
[29] Simon, H.M., Dodsworth, J.A. and Goodman, R.M. (2000) Cren-archaeota colonize terrestrial plant roots. Environ. Microbiol. 2, 495^ 505.
[30] Chelius, M.K. and Triplett, E.W. (2001) The diversity of Archaea and Bacteria in association with the roots of Zea mays L. Microb. Ecol. 41, 252^263.
[31] Bundt, M., Widmer, F., Pesaro, M., Zeyer, J. and Blaser, P. (2001) Preferential £ow paths : biological ‘hot spots’ in soils. Soil Biol. Bio-chem. 33, 729^738.
[32] United States Department of Agriculture (1994) Keys to soil taxon-omy, 6th edn., 524 pp. Pocahontas Press, Blacksburg, Virginia. [33] Flury, M. and Flu«hler, H. (1994) Brilliant Blue FCF as a dye tracer
for solute transport studies ^ a toxicological overview. J. Environ. Qual. 23, 1108^1112.
[34] Chatzinotas, A., Sandaa, R.-A., Scho«nhuber, W., Amann, R., Daae, F.L., Torsvik, V., Zeyer, J. and Hahn, D. (1998) Analysis of broad-scale di¡erences in microbial community composition of two pristine forest soils. Syst. Appl. Microbiol. 21, 579^587.
[35] Widmer, F., Seidler, R.J. and Watrud, L.S. (1996) Sensitive detection of transgenic plant marker gene persistence in soil microcosms. Mol. Ecol. 5, 603^613.
[36] Sandaa, R.-A., Enger, Y. and Torsvik, V. (1998) Rapid method for £uorometric quanti¢cation of DNA in soil. Soil Biol. Biochem. 30, 265^268.
[37] Widmer, F., Seidler, R.J., Gillevet, P.M., Watrud, L.S. and Di Gio-vanni, G.D. (1998) A highly selective PCR protocol for detecting 16S rRNA genes of the genus Pseudomonas (sensu stricto) in environmen-tal samples. Appl. Environ. Microbiol. 64, 2545^2553.
[38] Amann, R.I., Ludwig, W. and Schleifer, K. (1995) Phylogenetic iden-ti¢cation and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59, 143^169.
[39] Widmer, F., Sha¡er, B.T., Porteous, L.A. and Seidler, R.J. (1999) Analysis of nifH gene pool complexity in soil and litter at a douglas
¢r forest site in the Oregon cascade mountain range. Appl. Environ. Microbiol. 65, 374^380.
[40] Widmer, F., Fliessbach, A., Lacko¤, E., Schulze-Aurich, J. and Zeyer, J. (2001) Assessing soil biological characteristics : a comparison of bulk soil community DNA-, PLFA-, and Biolog1-analyses. Soil Biol. Biochem. 33, 1029^1036.
[41] Raaijmakers, J.G.W. (1987) Statistical analysis of the Michaelis^ Menten equation. Biometrics 43, 793^803.
[42] Hall, T.A. (1999) BioEdit : a user-friendly biological sequence align-ment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95^98.
[43] Felsenstein, J. (1989) PHYLIP ^ Phylogeny Inference Package (Ver-sion 3.2). Cladistics 5, 164^166.
[44] Jukes, T.H. and Cantor, C.R. (1969) Evolution of protein molecules. In: Mammalian Protein Metabolism (Munro, H.N. Eds.), pp. 21^ 132. Academic Press, New York.
[45] Saitou, N. and Nei, M. (1987) The neighbour-joining method : a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406^425.
[46] Olsen, G.J., Matsuda, H., Hagstrom, R. and Overbeel, R. (1994) FastDNAml : a tool for construction of phylogenetic trees of DNA sequences using maximum likelihood. Comput. Appl. Biosci. 10, 41^ 48.
[47] Kumar, S., Tamura, K., Jakobsen, I., Nei, M. (2001) MEGA2 : Mo-lecular Evolutionary Genetics Analysis Software 17, pp. 1244^1245. [48] Takai, K., Moser, D.P., DeFlaun, M., Onstott, T.C. and
Fredrick-son, J.K. (2001) Archaeal diversity in waters from deep South Afri-can gold mines. Appl. Environ. Microbiol. 67, 5750^5760. [49] Dojka, M.A., Hugenholtz, P., Haack, S.K. and Pace, N.R. (1998)
Microbial diversity in a hydrocarbon- and
chlorinated-solvent-con-taminated aquifer undergoing intrinsic bioremediation. Appl. Envi-ron. Microbiol. 64, 3869^3877.
[50] Kudo, Y., Shibata, S., Miyaki, T., Aono, T. and Oyaizu, H. (1997) Peculiar Archaea found in Japanese paddy soils. Biosci. Biotechnol. Biochem. 61, 917^920.
[51] Chin, K.J., Lukow, T. and Conrad, R. (1999) E¡ect of temperature on structure and function of the methanogenic archaeal community in an anoxic rice ¢eld soil. Appl. Environ. Microbiol. 65, 2341^ 2349.
[52] Zepp-Falz, K., Holliger, C., Grosskopf, R., Liesack, W., Nozhevni-kova, A.N., Muller, B., Wehrli, B. and Hahn, D. (1999) Vertical distribution of methanogens in the anoxic sediment of Rotsee (Swit-zerland). Appl. Environ. Microbiol. 65, 2402^2408.
[53] Wachinger, G., Fiedler, S., Zepp, K., Gattinger, A., Sommer, M. and Roth, K. (2000) Variability of soil methane production on the micro-scale : spatial association with hot spots of organic material and archaeal populations. Soil Biol. Biochem. 32, 1121^1130.
[54] Stein, J.L., Marsh, T.L., Wu, K.Y., Shizuya, H. and DeLong, E.F. (1996) Characterization of uncultivared prokaryotes : isolation and analysis of a 40-kilobase-pair genome fragment from a planktonic marine Archaeon. J. Bacteriol. 178, 591^599.
[55] Schleper, C., DeLong, E.F., Preston, C.M., Feldman, R.A., Wu, K.Y. and Swanson, R.V. (1998) Genomic analysis reveals chromo-somal variation in natural populations of the uncultured psychro-philic Archaeon Cenarchaeum symbiosum. J. Bacteriol. 180, 5003^ 5009.
[56] Schleper, C., Swanson, R.V., Mathur, E.J. and DeLong, E.F. (1997) Characterization of a DNA polymerase from the uncultivated psy-chrophilic archaeon Cenarchaeum symbiosum. J. Bacteriol. 179, 7803^ 7811.
